Search bioRxiv⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/043158

A method for downstream analysis of gene set enrichment results facilitates the biological interpretation of vaccine efficacy studies

Yan Tan·Jernej Godec·Felix Wu·Pablo Tamayo·Jill P Mesirov·W Nicholas Haining
View original source ↗Export metadata ↓

Abstract

Gene set enrichment analysis (GSEA) is a widely employed method for analyzing gene expression profiles. The approach uses annotated sets of genes, identifies those that are coordinately up- or down-regulated in a biological comparison of interest, and thereby elucidates underlying biological processes relevant to the comparison. As the number of gene sets available in various collections for enrichment analysis has grown, the resulting lists of significant differentially regulated gene sets may also become larger, leading to the need for additional downstream analysis of GSEA results. Here we present a method that allows the rapid identification of a small number of co-regulated groups of genes - \"leading edge metagenes\" (LEMs) - from high scoring sets in GSEA results. LEM are sub-signatures which are common to multiple gene sets and that \"explain\" their enrichment specific to the experimental dataset of interest. We show that LEMs contain more refined lists of context-dependent and biologically meaningful genes than the parental gene sets. LEM analysis of the human vaccine response using a large database of immune signatures identified core biological processes induced by five different vaccines in datasets from human peripheral blood mononuclear cells (PBMC). Further study of these biological processes over time following vaccination showed that at day 3 post-vaccination, vaccines derived from viruses or viral subunits exhibit patterns of biological processes that are distinct from protein conjugate vaccines; however, by day 7 these differences were less pronounced. This suggests that the immune response to diverse vaccines eventually converge to a common transcriptional response. LEM analysis can significantly reduce the dimensionality of enriched gene sets, improve the identification of core biological processes active in a comparison of interest, and simplify the biological interpretation of GSEA results.\n\nAuthor SummaryGenome-wide expression profiling is a widely used tool to identify biological mechanisms in a comparison of interest. One analytic method, Gene set enrichment analysis (GSEA) uses annotated sets of genes and identifies those that are coordinately up- or down-regulated in a biological comparison of interest. This approach capitalizes on the fact that alternations in biological processes often cause the coordinated change of a large number of genes. However, as the number of gene sets available in various collections for enrichment analysis has grown, the resulting lists of significant differentially regulated gene sets may also become larger, leading to the need for additional downstream analysis of GSEA results. Here we present a method that allows the identification of a small number of co-regulated groups of genes - \"leading edge metagenes\" (LEMs) - from high scoring sets in GSEA results. We show that LEMs contain more refined lists of context-dependent biologically meaningful genes than the parental gene sets and demonstrate the utility of this approach in analyzing the transcriptional response to vaccination. LEM analysis can significantly reduce the dimensionality of enriched gene sets, improve the identification of core biological processes active in a comparison of interest, and facilitate the biological interpretation of GSEA results.

Explore related subjects

BioinformaticsPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Yan Tan, Jernej Godec, Felix Wu, Pablo Tamayo, Jill P Mesirov, W Nicholas Haining. 2016-04-11. A method for downstream analysis of gene set enrichment results facilitates the biological interpretation of vaccine efficacy studies. https://doi.org/10.1101/043158

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2019

Detection and assembly of novel sequence insertions using Linked-Read technology

MotivationEmerging Linked-Read (aka read-cloud) technologies such as the 10x Genomics Chromium system have great potential for accurate detection and phasing of largescale human genome structural variations (SVs). By leveraging the long-range information encoded in Linked-Read sequencing, computational techniques are able to detect and characterize complex structural variations that are previously undetectable by short-read methods. However, there is no available Linked-Read method for detection and assembly of novel sequence insertions, DNA sequences present in a given sequenced sample but missing in the reference genome, without requiring whole genome de novo assembly. In this paper, we propose a novel integrated alignment-based and local-assembly-based algorithm, Novel-X, that effectively uses the barcode information encoded in Linked-Read sequencing datasets to improve detection of such events without the need of whole genome de novo assembly. We evaluated our method on two haploid human genomes, CHM1 and CHM13, sequenced on the 10x Genomics Chromium system. These genomes have been also characterized with high coverage PacBio long-reads recently. We also tested our method on NA12878, the wellknown HapMap CEPH diploid genome and the child genome in a Yoruba trio (NA19240) which was recently studied on multiple sequencing platforms. Detecting insertion events is very challenging using short reads and the only viable available solution is by long-read sequencing (e.g. PabBio or ONT). Our experiments, however, show that Novel-X finds many insertions that cannot be found by state of the art tools using short-read sequencing data but present in PacBio data. Since Linked-Read sequencing is significantly cheaper than long-read sequencing, our method using Linked-Reads enables routine large-scale screenings of sequenced genomes for novel sequence insertions. AvailabilitySoftware is freely available at https://github.com/1dayac/novel_insertions Contactimh2003@med.cornell.edu Supplementary informationSupplementary data are available at https://github.com/1dayac/novel_insertions_supplementary

Bioinformatics↗
bioRxiv2016

iCARE: An R Package to Build and Apply Absolute Risk Models

This report describes a R package, called the Individualized Coherent Absolute Risk Estimation (iCARE) tool, that allows researchers to build and evaluate models for absolute risk and apply them to estimate an individuals risk of developing disease during a specified time interval based on a set of user defined input parameters. An attractive feature of the software is that it gives users flexibility to update models rapidly based on new knowledge on risk factors and tailor models to different populations by specifying three input arguments: (1) a model for relative risk, (2) an age-specific disease incidence rate, (3) the distribution of risk factors for the population of interest. The tool can handle missing information on risk factors for individuals for whom risks are to be predicted using a coherent approach where all estimates are derived from a single model after appropriate model averaging. The software allows single nucleotide polymorphisms (SNPs) to be incorporated into the model using published odds ratios and allele frequencies. The validation component of the software implements the methods for evaluation of model calibration, discrimination and risk-stratification based on independent validation datasets. We provide an illustration of the utility of iCARE for building, validating and applying absolute risk models using breast cancer as an example.

Bioinformatics↗
bioRxiv2016

deSPI: efficient classification of metagenomic reads with lightweight de Bruijn graph-based reference indexing

SummaryIn metagenomic studies, fast and effective tools are on wide demand to implement taxonomy classification for upto billions of reads. Herein, we propose deSPI, a novel read classification method that classifies reads by recognizing and analyzing the matches between reads and reference with de Bruijn graph-based lightweight reference indexing. deSPI has faster speed with relatively small memory footprint, meanwhile, it can also achieve higher or similar sensitivity and accuracy.\n\nAvailabilitythe C++ source code of deSPI is available at https://github.com/hitbc/deSPI\n\nContactydwang@hit.edu.cn\n\nSupplementary informationSupplementary data are available at Bioinformatics online.

Bioinformatics↗